Power equipment state evaluation method based on data mining and principal component analysis

Abstract

The invention discloses a power equipment state evaluation method based on data mining and principal component analysis (PCA). The method comprises the following steps: determining a parameter to be evaluated of power equipment; mining and analyzing the determined parameter to be evaluated of the power equipment, establishing a comprehensive evaluation module for the operating state of the power equipment by using a PCA method; selecting modeling data, solving model parameters, determining an early warning value, and performing real-time dynamic evaluation. The method establishes a power equipment key parameter system by using dimensionality reduction technology, reduces the computation space and time, improves operation rate and efficiency; and can achieve differentiated evaluation on different individuals of the same kind of equipment in the case of lack of historical failure sample data of the power equipment.

Description

technical field [0001] The invention relates to the technical field of power equipment state monitoring, in particular to a power equipment state evaluation method based on data mining and principal component analysis. Background technique [0002] The normal operation of power equipment is the guarantee of the safety and stability of the power system, so the state evaluation and fault prediction of power equipment are particularly important. [0003] There are many problems in the current state evaluation research of power equipment: there are many state indicators of power equipment, but there is no comprehensive evaluation system, which leads to insufficient utilization of index information and incomplete state evaluation; most of the existing methods need to use equipment failure data, but the actual There are few cases of equipment failure on the Internet; when the threshold method is used to evaluate similar equipment, it cannot reflect the individual differences of th...

AI Technical Summary

Problems solved by technology

[0003] There are many problems in the current state evaluation research of power equipment: there are many state indicators of power equipment, but there is no comprehensive evaluation system, which leads to insufficient utilization of index information and incomplete state evaluation; most of the existing methods need to use equipment failure data, but the actual Lack of equipment failure cases on the Internet; when using the threshold method to evaluate similar equipment, it cannot reflect the individual differences of the equipment, the accuracy is too low, etc.

[0008] First, in this patent, only the eigenvalues ​​in the direction of the first principal component are used, and the eigenvalues ​​in the direction of the second, third, fourth, etc. are omitted. The eigenvalues ​​in the direction of the first principal component can only express The information contribution rate of the first principal component cannot fully represent the characteristics of all analyzed variables, and the information of the original variables contained in the second, third, and fourth principal components is omitted, which increases the judgment error

[0009] Second, in this patent, the eigenvalue in the direction of the first principal component is selected as the state evaluation value, which is not comprehensive enough and can only reflect the changing state of the first principal component

Images